def pull_function(**context): user_id = context['ti'].xcom_pull(task_ids='push_task', key='user_id') print(f"Received user_id")
XComs are for coordination , not data transfer . Final Takeaway XComs are Airflow’s glue. They turn a set of isolated tasks into a coherent pipeline. Use them for small control signals, IDs, and results. Keep them light. And when you’re tempted to pass a big blob of data – stop, and ask yourself: should this be in object storage instead? xcom in airflow
@task def consume_two(data): return f"Got data['source']" @task def fetch_urls() -> list[str]: return ["http://a.com", "http://b.com"] @task def download(url: str) -> str: # download content return f"content_of_url" def pull_function(**context): user_id = context['ti']
process_record(get_latest_record_id()) @task def produce_data(): return "ids": [1,2,3], "source": "api" @task def consume_one(data): return f"Got data['ids'][0]" Use them for small control signals, IDs, and results
No xcom_push or xcom_pull needed – the TaskFlow wiring handles it. With traditional operators, you must push/pull manually.
push = PythonOperator(task_id='push_task', python_callable=push_function) pull = PythonOperator(task_id='pull_task', python_callable=pull_function)
from airflow.operators.python import PythonOperator def push_function(**context): context['ti'].xcom_push(key='user_id', value=123)